Needle in a Haystack: Tracking UAVs from Massive Noise in Real-World 5G-A Base Station Data

TL;DR

This paper presents BSense, a novel system that tracks UAVs using commercial 5G-A base station point clouds, effectively filtering noise and false detections to achieve high accuracy in urban environments.

Contribution

We introduce a layered filtering framework leveraging signal fingerprints, spatial and velocity consistency, and Transformer-based motion pattern analysis for UAV tracking.

Findings

Reduces false detections from 168.05 to 0.04 per frame

Achieves 95.56% F1 score in urban UAV tracking

Localizes UAVs with an average error of 4.9 meters at 1,000 meters range

Abstract

The potential usage of UAVs in daily life has made monitoring them essential. However, existing systems for monitoring UAVs typically rely on cameras, LiDARs, or radars, whose limited sensing range or high deployment cost hinder large-scale adoption. In response, we develop BSense, the first system that tracks UAVs by leveraging point clouds from commercial 5G-A base stations. The key challenge lies in the dominant number of noise points that closely resemble true UAV points, resulting in a noise-to-UAV ratio over 100:1. Therefore, identifying UAVs from the raw point clouds is like finding a needle in a haystack. To overcome this, we propose a layered framework that filters noise at the point, object, and trajectory levels. At the raw point level, we observe that noise points from different spatial regions exhibit distinguishable and consistent signal fingerprints, which we can model to identify and remove them. At the object level, we design spatial and velocity consistency checks to identify false objects, and further compute confidence scores by aggregating these checks over multiple frames for more reliable discrimination. At the final trajectory level, we propose a Transformer-based network that captures multi-frame motion patterns to filter the few remaining false trajectories. We evaluated BSense on a commercial 5G-A base station deployed in an urban environment. The UAV was instructed to fly along 25 distinct trajectories across 54 cases over 7 days, yielding 155 minutes of data with more than 14,000 frames. On this dataset, our system reduces the number of false detections from an average of 168.05 per frame to 0.04, achieving an average F1 score of 95.56% and a mean localization error of 4.9 m at ranges up to 1,000 m.